Problematic sexual behaviors as a juvenile outcome measure

By Norbert Ralph, PhD, MPH

 

For juveniles who sexually offended (JwSO), sexual recidivism is identified as one of the primary outcome measures. Reitzel and Carbonell (2006) titled their meta-analysis of treatment programs "The Effectiveness of Sexual Offender Treatment for Juveniles as Measured by Recidivism." Methodologically and clinically using this measure has virtues but also limitations. With sexual recidivism rates in some researchers estimated to be about 5% (Lussier et al., 2023) it would be difficult to obtain samples large enough to detect a treatment effect. Also would a reduction from 5% to even 0% be significant enough in terms of real-world impacts and cost/benefit considerations?

 

Other outcomes have been used. Nonsexual recidivism has been identified as another benchmark to assess treatment outcomes. Lussier et al. (2023) identified general recidivism for this population as 44% and Caldwell (2016) as 27% in meta-analytic studies. General recidivism has been a focus of study, and for example, Abei et al. (2022) carried out a study of high-risk JwSO youth to compare the efficacy of sexual offense-oriented therapy and social skills training in the prevention of both sexual and general recidivism. Further, the Multisystemic Therapy group used out-of-home placements, among other measures (Borduin & Munschy, 2021).

 

Another benchmark for juveniles who sexually offended is examining problematic sexual behaviors (PSB) that may not result in arrests or formal recidivism. Viljoen et al. (2007) reported a rate of such behaviors in a residential treatment program with 169 JwSO youth of 16.6% and an average JSORRAT-II score of 6.1. Ralph (2015) in a study of another residential treatment program with 129 JwSO youth reported a rate of PSB of 20.6% and an average JSORRAT-II score of 6.3. The average JSORRAT-II scores (6.1 and 6.3 respectively) of the samples indicated that both groups above the average risk levels (Epperson, 2019). In the latter study (Ralph, 2015), any sexual behavior that violated the rules of the setting was classified as misbehavior. Notably none of these behaviors resulted in charges, even though some were serious enough to be charged. Presumably, this was considered not necessary because these youth were already on probation for such offenses and were in court-ordered treatment. For context is important to note findings such as Ybarra & Mitchell (2013) which identified nearly 1 in 10 youths (9%) reported some type of sexual violence perpetration in their lifetime.

 

Another example including PSB is in Letourneau et al. (2009), which reported an outcome study regarding Multisystemic Therapy (MST) using the Adolescent Sexual Behavior Inventory (Friedrich, Lysne, Sims, & Shamos, 2004) and its Sexual Risk/Misuse subscale. Because of the nature of the scale, specific PSBs weren't possible to separate out, such as coercing others to have sex. Abei et al. (2022) in the study mentioned previously, classified as sexual recidivism not only formal charges but also PSB that may have resulted in a formal charge but did not. However, the rate of such PSB separate from charged offenses was not separately described.

 

The above information has several implications regarding assessment and treatment for these youth. One is the recommendation that such behaviors be included in outcome studies assessing treatment effectiveness and also for individual programs as part of their quality assurance procedures. These behaviors are important to address since they may have victims too and likely increase the risk of future such behaviors. Once identified, these behaviors are more likely to become an explicit focus of treatment. Notably, the Abei et al. (2022) study has already included PSB. Also, PSB may be tracked systematically in future MST studies

 

There is another consideration. It might be possible to develop risk measures to assess for PSB that do not result in formal charges. For example, in the study by Ralph (2015) youth examining PSB, who had a male victim and previous mental health treatment had an AUC (Area Under the Curve) of .74. Practically, developing risk measures to assess for PSB might be more possible than developing measures to assess sexual recidivism alone, given the low sexual recidivism rates. Clinically, such measures would be useful to assess the risk of a given youth for sexual acting out at the beginning of treatment, particularly for high-risk youth, so that treatment methods and the amount of treatment could adequately be planned.